Diamond Quantum Memory Beats Record

It holds a quantum bit long enough for real computation

3 min read

7 June 2012—A new type of memory could give the emerging field of quantum computing a much-needed boost. Scientists at Harvard University say they’ve developed a solid-state, room-temperature quantum memory that can hold information longer than one second. The memory records a quantum property called spin on the nucleus of an atom inside a diamond. Earlier solid-state nuclear-spin-based memories would lose their data after only milliseconds unless cryogenically cooled. The research is being reported in this week’s issue of the journal Science.  

For practical quantum computers—experimental machines that aim to solve problems beyond the reach of ordinary computers by exploiting some of the stranger rules of quantum mechanics—it is essential to have quantum memories that last long enough to process information and churn out answers. Physicists think that the needed time frame is about one second.

“Diamond turns out to be a particularly good material for [the] purpose,” says Peter Maurer, a graduate student in physics at Harvard who came up with the new memory along with his advisor, Mikhail Lukin, and other colleagues. “It is possible to fabricate diamond with very few impurities.”

The Harvard scientists exploited such an impurity, a defect in the diamond lattice known as a nitrogen vacancy, to create a quantum bit, or qubit. The qubit could interact with a nearby carbon-13 atom in the diamond lattice, which acted as the quantum memory.  

The spin of an electron at the defect can be manipulated by pulses from a laser, making it a qubit. The spin at the center of the nitrogen vacancy was thus made to act as the control qubit, which was used to manipulate the spin of the nucleus of the nearby carbon-13 atom. Although the interaction between the spins was weak, it proved strong enough to create a quantum memory that stored information for a full second.

The general technique has been around for a while. “People have been using nitrogen-vacancy centers as quantum registers,” Maurer says. “However, the coherence times were always limited to a few milliseconds.” Coherence is the term physicists use to denote how long the quantum state remains pure and stable. Quantum states are notoriously finicky and soon “decohere,” or interact with their environments, losing whatever quantum information they carried. The challenge of building quantum memories is to maintain coherence long enough to make useful calculations.

Maurer says the key was finding a way to use the laser to keep the electron from decohering, allowing the researchers to extend the coherence time past one second.

That time, says Maurer, is three orders of magnitude better than previously attained using this method. It is long enough to be useful for the quantum repeaters needed for an unhackable communication network, quantum computation schemes, and quantum money (money that in theory cannot be counterfeited).

Maurer says there is already interest in the team’s approach. “Our experimental techniques are being patented, and a U.K.-based company has expressed interest in licensing our idea,” he says.

Experts are impressed, particularly because these quantum memories do not require cooling. Most techniques require cryogenic temperatures, which could pose a barrier to their commercialization.

Physicist Dane McCamey of the University of Sydney, in Australia, who wrote an accompanying perspective in Science , calls it “an important achievement.”

“Technologically, it should allow them to implement error-correction schemes similar to those used in dynamic RAM, increasing the likelihood of successfully engineering large-scale quantum information-processing systems based in nitrogen-vacancy centers in diamond, which operate at room temperature,” McCamey says.

He adds that the new technique may well be able to store quantum information for hours, or even days, in a portable, room-temperature solid-state device.

“This work challenges widely held beliefs that quantum systems need to be cold to have useful lifetimes and provides a pathway toward the exploitation of coherent quantum mechanical systems in everyday life,” McCamey says.

About the Author

Saswato R. Das, a New York City–based writer, contributes frequently to IEEE Spectrum . In April, he reported on the creation of a two-node quantum network that could form the basis of a quantum internet.

The Conversation (0)

The Future of Deep Learning Is Photonic

Computing with light could slash the energy needs of neural networks

10 min read

This computer rendering depicts the pattern on a photonic chip that the author and his colleagues have devised for performing neural-network calculations using light.

Alexander Sludds

Think of the many tasks to which computers are being applied that in the not-so-distant past required human intuition. Computers routinely identify objects in images, transcribe speech, translate between languages, diagnose medical conditions, play complex games, and drive cars.

The technique that has empowered these stunning developments is called deep learning, a term that refers to mathematical models known as artificial neural networks. Deep learning is a subfield of machine learning, a branch of computer science based on fitting complex models to data.

Keep Reading ↓ Show less